. .. Conception,

.. .. Émotions,

. .. Conclusion,

. .. Et-indexation, 72 7.2 Détection automatique de non-adhérence, Corpus : collecte, annotation

, Chapitre 8

B. Bibliographie, «Guidelines for atc classification and ddd assignment», cahier de recherche, WHO Collaborating Centre for Drug Statistics Methodology, 2019.

L. Aagaard, J. Strandell, L. Melskens, P. Petersen, E. Holme et al., «Global patterns of adverse drug reactions over a decade : analyses of spontaneous reports to VigiBase?, Drug Saf, vol.35, issue.3, pp.1171-82, 2012.

R. Abdellaoui, P. Foulquié, N. Texier, C. Faviez, A. Burgun et al., «Detection of Cases of Noncompliance to Drug Treatment in Patient Forum Posts : Topic Model Approach», Journal of Medical Internet Research, vol.20, issue.3, p.85, 2018.

E. Arsevska, M. Roche, P. Hendrikx, D. Chavernac, S. Falala et al., Identification of terms for detecting early signals of emerging infectious disease outbreaks on the web», Computers and Electronics in Agriculture, vol.123, pp.104-115, 2016.

S. Ayvaz, J. Horn, O. Hassanzadeh, Q. Zhu, J. Stan et al., «Toward a complete dataset of drug-drug interaction information from publicly available sources, J Biomed Inform, vol.55, issue.3, pp.206-223, 2015.

A. Bate, M. Lindquist, I. Edwards, S. Olsson, R. Orre et al., «A bayesian neural network method for adverse drug reaction signal generation», Eur J Clin Pharmacol, vol.54, issue.4, pp.315-321, 1998.

F. Benamara, V. Moriceau, J. Mothe, F. Ramiandrisoa, and Z. He, Automatic detection of depressive users in social media, 2018.

C. Bousquet, G. Lagier, A. Lillo-le-louët, C. Le, A. Beller et al., Appraisal of the meddra conceptual structure for describing and grouping adverse drug reactions, vol.28, pp.19-34, 2005.

L. Breiman, Machine Learning, vol.45, pp.5-32, 2001.

P. Brown, P. Desouza, R. Mercer, V. Della-pietra, and J. L. , «Class-based n-gram models of natural language, Computational Linguistics, vol.18, issue.4, pp.467-479, 1992.

D. Cameron, G. A. Smith, R. Daniulaityte, A. P. Sheth, D. Dave et al., «Predose : A semantic web platform for drug abuse epidemiology using social media, Journal of Biomedical Informatics, vol.46, issue.5, pp.985-997, 2013.

H. S. Campbell, M. R. Phaneuf, and K. D. , «Cancer peer support programs-do they work ?», Patient education and counseling, vol.55, pp.3-15, 2004.

T. P. Center, The social life of health information, 2011.

N. Collier, «Towards cross-lingual alerting for bursty epidemic events, J Biomed Semantics, vol.2, issue.5, pp.10-14, 2011.

J. Colloc, Santé et big data : l'etat et les individus, impuissants face aux pouvoirs des réseaux», L'Espace Politique, vol.26, 2015.

T. Daugherty, M. S. Eastin, and L. Bright, Exploring consumer motivations for creating usergenerated content, Journal of Interactive Advertising, vol.8, issue.2, pp.16-25, 2008.

J. L. Donovan and D. R. Blake, «Patient non-compliance : deviance or reasoned decision-making ?», Social science & medicine, vol.34, pp.507-513, 1992.

S. Duda, C. Aliferis, R. Miller, A. Slatnikov, and K. J. , «Extracting drug-drug interaction articles from Medline to improve the content of drug databases, dans Ann Symp Am Med Inform Assoc (AMIA), pp.216-236, 2005.

E. Ekman and M. Krasner, Empathy in medicine : Neuroscience, education and challenges, vol.39, pp.164-173, 2017.

M. Feehan, M. A. Morrison, C. Tak, D. E. Morisky, M. M. Deangelis et al., «Factors predicting self-reported medication low adherence in a large sample of adults in the US general population : a cross-sectional study, BMJ Open, vol.7, issue.6, 2017.

J. L. Fleiss and J. Cohen, «The equivalence of weighted kappa and the intraclass correlation coefficient as measures of reliability, Educational and Psychological Measurement, vol.33, pp.613-619, 1973.

N. Gauducheau, La communication des émotions dans les échanges médiatisés par ordinateur : bilan et perspectives», vol.61, pp.389-404, 2008.

L. Goeuriot, J. Na, W. Y. Kyaing, C. S. Khoo, Y. L. Theng et al., Textual and informational characteristics of drug-related content on three kinds of websites : Drug review website, discussion board and hospital information portal, vol.2, pp.27-49, 2011.

H. White and M. S. Dorman, Online support for caregivers : Analysis of an internet alzheimer mailgroup», Computers in nursing, vol.18, pp.168-76, 1999.

J. Halpern, «What is clinical empathy ?», Journal of general internal medicine, vol.18, issue.3, pp.670-674, 2003.

T. Hamon and N. Grabar, Extraction of ingredient names from recipes by combining linguistic annotations and crf selection», dans Proceedings of the 5th International Workshop on Multimedia for Cooking &#38, Eating Activities, CEA '13, pp.63-68, 2013.

R. B. Haynes, H. Mcdonald, A. X. Garg, and P. Montague, «Interventions for helping patients to follow prescriptions for medications, The Cochrane Database of Systematic Reviews, vol.1, issue.2, pp.0-011, 2002.

R. Hipp and . Sqlite, , 2000.

M. Hojat, D. Louis, F. Markham, R. Wender, C. Rabinowitz et al., «Physicians' empathy and clinical outcomes for diabetic patients», Academic medicine : journal of the Association of American Medical Colleges, vol.86, pp.359-64, 2011.

J. Hugtenburg, L. Timmers, P. Elders, M. Vervloet, and L. Van-dijk, «Definitions, variants, and causes of nonadherence with medication : A challenge for tailored interventions, Patient preference and adherence, vol.7, pp.675-82, 2013.

J. G. Hugtenburg, L. Timmers, P. J. Elders, M. Vervloet, and L. Van-dijk, «Definitions, variants, and causes of nonadherence with medication : a challenge for tailored interventions, Patient preference and adherence, vol.7, pp.675-682, 2013.

G. H. John and P. Langley, «Estimating continuous distributions in bayesian classifiers», dans Eleventh Conference on Uncertainty in Artificial Intelligence, édité par M. Kaufmann, pp.338-345, 1995.

J. Kalyanam, T. Katsuki, G. R. Lanckriet, and T. K. Mackey, Exploring trends of nonmedical use of prescription drugs and polydrug abuse in the twittersphere using unsupervised machine learning, Addictive Behaviors, vol.65, pp.289-295, 2017.

C. Lacoste-roussillon, P. Pouyanne, F. Haramburu, G. Miremont, and B. Bégaud, «Incidence of serious adverse drug reactions in general practice : a prospective study, Clin Pharmacol Ther, vol.69, issue.6, pp.458-462, 2001.

M. Lafourcade, Making people play for lexical acquisition, dans Symp on Natural Language Processing, vol.26, 2007.
URL : https://hal.archives-ouvertes.fr/lirmm-00200883

N. Landwehr, M. Hall, and E. F. , «Logistic model trees, Machine Learning, vol.95, pp.161-205, 2005.

G. Lejeune, R. Brixtel, C. Lecluze, A. Doucet, and N. L. , Added-value of automatic multilingual text analysis for epidemic surveillance, Artificial Intelligence in Medicine, p.4, 2013.
URL : https://hal.archives-ouvertes.fr/hal-01074535

P. Liang, Semi-Supervised Learning for Natural Language, Master, Massachusetts Institute of Technology, p.27, 2005.

F. Liebens, M. Aimont, B. Carly, A. Pastijn, S. Swimberg et al., «Internet, presse, médias : nouveaux éléments dans la communication médicale», 27°Journées de la Société française de sénologie et de pathologie mammaire, Dogmes et doutes (revue critique des standards en sénologie)/Dogmas and doubts, p.4, 2005.

M. Marcoccia, «L'animation d'un espace numérique de discussion : L'exemple des forums usenet», Document numérique, vol.5, 2001.

Z. A. Marcum, M. A. Sevick, and S. M. Handler, «Medication Nonadherence, vol.309, pp.2105-2106, 2013.

A. Mccallum and K. Nigam, «A comparison of event models for naive bayes text classification, dans AAAI workshop on Learning for Text Categorization, p.36, 1998.

C. A. Mchorney and C. V. Spain, Frequency of and reasons for medication non-fulfillment and non-persistence among american adults with chronic disease in, Health Expectations, vol.14, pp.307-320, 2008.

S. Melzi, A. Abdaoui, J. Azé, S. Bringay, P. Poncelet et al., «Patient's rationale : Patient Knowledge retrieval from health forums», dans eTELEMED : eHealth, 2014.

T. Mikolov, K. Chen, G. Corrado, and J. Dean, «Efficient estimation of word representations in vector space, p.28, 2013.

T. Mikolov, I. Sustkever, K. Chen, G. Corrado, and J. Dean, Distributed representations of words and phrases and their compositionality», dans NIPS, p.28, 2013.

S. J. Moon, W. Lee, J. S. Hwang, Y. P. Hong, and D. E. Morisky, Accuracy of a screening tool for medication adherence : A systematic review and meta-analysis of the Morisky Medication Adherence Scale-8», PLoS ONE, vol.12, p.11, 2017.

Y. Moride, F. Haramburu, A. R. Alvarez, and B. Bégaud, Under-reporting of adverse drug reactions in general practice, Br J Clin Pharmacol, vol.43, issue.2, pp.177-181, 1997.

F. Morlane-hondère, C. Grouin, and P. Zweigenbaum, «Identification of drug-related medical conditions in social media», dans LREC, pp.1-7, 2016.

H. Nabarette, «L'internet medical et la consommation d'information par les patients, vol.114, 2002.

N. Natarajan, W. Putnam, K. Van-aarsen, K. Lawson, and F. Burge, «Adherence to antihypertensive medications among family practice patients with diabetes mellitus and hypertension, Canadian Family Physician Medecin De Famille Canadien, vol.59, issue.2, pp.93-100, 2013.

M. Neumann, F. Edelhäuser, D. Tauschel, M. Fischer, M. Wirtz et al., Empathy decline and its reasons : a systematic review of studies with medical students and residents, 2011.

A. Nikfarjam, A. Sarker, K. O'connor, R. Ginn, and G. Gonzalez, «Pharmacovigilance from social media : mining adverse drug reaction mentions using sequence labeling with word embedding cluster features, Journal of the American Medical Informatics Association, vol.22, pp.671-681, 2015.

K. O'connor, P. Pimpalkhute, A. Nikfarjam, R. Ginn, K. L. Smith et al., Pharmacovigilance on twitter ? mining tweets for adverse drug reactions, pp.924-933, 2014.

. Oms, Classification statistique internationale des maladies et des problèmes de santé connexes -Dixième révision, Organisation mondiale de la Santé, vol.19, 1995.

J. Platt, «Fast training of support vector machines using sequential minimal optimization, dans Advances in Kernel Methods -Support Vector Learning, édité par B. Schoelkopf, C. Burges et A. Smola, 1998.

R. Plutchik, The emotions, p.69, 1991.

P. Pouyanne, F. Haramburu, J. Imbs, and B. Bégaud, «Admissions to hospital caused by adverse drug reactions : cross sectional incidence study. French pharmacovigilance centres, BMJ, vol.320, issue.3, pp.1036-1036, 2000.

P. Queneau, B. Bannwarth, F. Carpentier, J. Guliana, J. Bouget et al., «Emergency department visits caused by adverse drug events : results of a French survey, Drug Saf, vol.30, issue.1, pp.81-88, 2007.

J. Quinlan, C4.5 Programs for Machine Learning, p.36, 1993.

H. Riess, «The impact of clinical empathy on patients and clinicians : Understanding empathy's side effects, AJOB Neuroscience, vol.6, pp.51-53, 2015.

S. Rodgers and Q. Chen, Internet community group participation : Psychosocial benefits for women with breast cancer, Journal of Computer-Mediated Communication, vol.10, pp.0-00, 2005.

H. Romeyer, Tics et santé : entre information médicale et information de santé, vol.10, p.12, 2008.

H. Romeyer, La santé en ligne, 2012.

A. Sarker, R. Ginn, A. Nikfarjam, K. O'connor, K. Smith et al., Utilizing social media data for pharmacovigilance : A review, J Biomed Inform, vol.54, pp.202-212, 2015.

H. Schmid, «Probabilistic part-of-speech tagging using decision trees, dans Int Conf on New Methods in Language Processing, vol.13, p.37, 1994.

I. Segura-bedmar, P. Martinez, and M. Herrero-zazo, SemEval-2013 task 9 : Extraction of drugdrug interactions from biomedical texts (DDIExtraction 2013)», dans Lexical and Computational Semantics (*SEM), pp.341-350, 2013.

Y. Shang, «Users' participation motivation and behavior patterns in online health community : A game theory viewpoint, 2019.

A. Simoës-perlant, T. Lanchantin, and P. Largy, «De la relation de l'écrit numérique et la qualité de l'orthographe, vol.118, pp.40-57, 2015.

A. Staii, «La fabrique sociale : autonomisation et légitimation dans le domaine de l'information de santé, vol.4, p.60, 2012.

. Statista and . Com, Classement des sites internet les plus visités depuis un ordinateur en france en décembre 2018, selon le nombre de visiteurs uniques par mois (en milliers), 2018.

T. Strohman, D. Metzler, H. Turtle, and W. B. Croft, «Indri : a language-model based search engine for complex queries, dans Proceedings of the International Conference on Intelligent Analysis, p.42, 2005.

M. Tapi-nzali, Analyse des médias sociaux de santé pour évaluer la qualité de vie des patientes atteintes d'un cancer du sein, 2017.

L. Townsend and C. W. , Social media research : A guide to ethics, 2016.

G. Trifirò, A. Pariente, P. Coloma, J. Kors, G. Polimeni et al., Eu-adr group. data mining on electronic health record databases for signal detection in pharmacovigilance : which events to monitor ?, Pharmacoepidemiol Drug Saf, vol.18, issue.3, pp.1176-84, 2009.

J. W. Turner, J. A. Grube, and J. Meyers, «Developing an optimal match within online communities : an exploration of cmc support communities and traditional support, Journal of Communication, vol.51, issue.2, pp.231-251, 2001.

O. Tw, D. Zb, A. Rm, and R. Ar, «Characteristics of physician empathetic statements during pediatric intensive care conferences with family members : A qualitative study, JAMA Network Open, vol.1, pp.180-351, 2018.

Y. Wang, R. Kraut, and J. Levine, Eliciting and receiving online support : using computer-aided content analysis to examine the dynamics of online social support, journal of medical Internet research, vol.17, issue.4, 2015.

J. N. Weinstein, K. Clay, and T. S. Morgan, Informed patient choice : patient-centered valuing of surgical risks and benefits, Health Affairs, vol.26, issue.3, pp.726-730, 2007.

, «Adherence to long-term therapies : evidence for action», cahier de recherche, vol.2, p.50, 2003.

I. Witten and E. F. , Data mining : Practical machine learning tools and techniques, p.34, 2005.

J. Xie, D. D. Zeng, and Z. A. Marcum, Using deep learning to improve medication safety : the untapped potential of social media, Therapeutic Advances in Drug Safety, vol.8, pp.375-377, 2017.

Z. Yan, T. Wang, Y. Chen, and H. Zhang, «Knowledge sharing in online health communities : A social exchange theory perspective, Information & Management, vol.53, pp.643-653, 2016.

J. Zhang, G. Le, D. Larochelle, R. Pasick, G. F. Sawaya et al., «Facts or stories ? how to use social media for cervical cancer prevention : A multi-method study of the effects of sender type and content type on increased message sharing, Preventive Medicine, vol.126, pp.105-751, 2019.

, Il peut s'agir d'une situation où le patient prend trop (sur-usage) ou pas assez (sous-usage) de médicaments, boit de l'alcool alors qu'il y a une contrindication, ou encore commet une tentative de suicide à l'aide de médicaments. Selon Haynes 2002améliorer l'adhérence pourrait avoir un plus grand impact sur la santé de la population que tout autre amélioration d'un traitement médical spécifique. Cependant les données sur la non-adhérence sont difficiles à acquérir, puisque les patients en situation de non-adhérence sont peu susceptibles de rapporter leurs actions à leurs médecins, Résumé La non-adhérence médicamenteuse désigne les situations où le patient ne suit pas les directives des autorités médicales concernant la prise d'un médicament

, Nous identifions 3 motivations : gérer soi-même sa santé, rechercher un effet différent de celui pour lequel le médicament est prescrit, être en situation d'addiction ou d'accoutumance. La gestion de sa santé recouvre ainsi plusieurs situations : éviter un effet secondaire, moduler l'effet du médicament, sous-utiliser un médicament perçu comme inutile, agir sans avis médical. Additionnellement, une non-adhérence peut survenir par erreur ou négligence, sans motivation particulière. À l'issue de notre étude nous produisons : un corpus annoté avec des messages de non-adhérence, Dans un premier temps, nous collectons un corpus de messages postés sur des forums médicaux. Nous construisons des vocabulaires de noms de médicaments et de maladies utilisés par les patients. Nous utilisons ces vocabulaires pour indexer les médicaments et maladies dans les messages

, According to Haynes 2002increasing drug compliance may have a bigger impact on public health than any other medical improvements. However non-compliance data are difficult to obtain since non-adherent patients are unlikely to report their behaviour to their healthcare providers. This is why we use data from social media to study drug non-compliance. Our study is applied to French-speaking forums. First we collect a corpus of messages written by users from medical forums. We build vocabularies of medication and disorder names such as used by patients. We use these vocabularies to index medications and disorders in the corpus. Then we use supervised learning and information retrieval methods to detect messages talking about non-compliance. With machine learning, we obtain 0.513 F-mesure, with up to 0.5 precision or 0.6 recall. With information retrieval we identify specific situations such as drinking contraindicated alcohol or using neuroleptics for their psychotropic effect. After that, we study the content of the non-compliance messages. We identify various non-compliance situations and patient's motivations. We identify 3 main motivations : self-medication, seeking an effect besides the effect the medication was prescribed for, or being in addiction or habituation situation. Selfmedication is an umbrella for several situations : avoiding an adverse effect, adjusting the medication's effect, underusing a medication seen as useless, taking decisions without a doctor's advice. Non-compliance can also happen thanks to errors or carelessness, without any particular motivation. Our work provides several kinds of result